Datasets:
| { | |
| "field_id": "F031", | |
| "field": "causal representation learning", | |
| "directions": [ | |
| { | |
| "direction_id": "G0", | |
| "name": "Learning latent causal structure and factors", | |
| "scope": "Learn latent causal variables, mechanisms, and their relationships from mixed observations, multiple views, environments, or intervention evidence. This includes both conditions for recoverability and recovery in practice, as well as learning interpretable, causally related factors with generative or factor models and testing their generative structure." | |
| }, | |
| { | |
| "direction_id": "G1", | |
| "name": "Causal representations for changing conditions", | |
| "scope": "Learn stable causal information or reusable mechanisms to support prediction, control, and model adaptation when environments, interventions, or goals change." | |
| }, | |
| { | |
| "direction_id": "G2", | |
| "name": "Intervention and counterfactual outcome modeling", | |
| "scope": "Build and evaluate causal generative models to estimate outcomes for specified intervention or counterfactual queries, or to generate samples conditioned on those queries." | |
| } | |
| ] | |
| } | |